This retrospective pre-post study compared patients admitted with a blood culture positive for a Gram-negative organism detectable by the BC-GN before (January 1, 2015 – July 1, 2015) and after (July 15, 2015 – January 15, 2016) this microarray test was implemented. The primary objective was to compare time from Gram stain to antimicrobial switch pre and post implementation. Secondary objectives included time from Gram stain to active treatment, in-hospital mortality, and hospital length of stay.
Results: 877 patients with Gram-negative bacteremia were included, 456 in the pre group and 421 in the post group. The most common organisms detected in the pre and post groups were E. coli (46.9% vs. 47.5%) and K. pneumoniae (19.1% vs. 20.9%). In the post group, 52 resistance genes were detected: 44 CTX-M, 6 KPC, and 2 OXA. The number and type of antimicrobial switches were similar between groups. Median (IQR) time from Gram stain to antimicrobial switch was significantly decreased in the post group, 28.6 (8.6 – 56.9) hr vs. 44.1 (18.9 – 64.6) hr, p = 0.0036. In patients on inactive antimicrobial therapy at the time of result, median time from Gram stain to active therapy was lower in the post group, 8.8 (5.5 – 18.4) hr vs. 24.5 (4.9 – 44.3) hr, p = 0.0344. Median (IQR) hospital length of stay was also decreased in the post group, 7 (5-15) days vs. 9 (4.5 – 21) days, p = 0.001. The rate of in-hospital mortality was similar between groups, 11.6% (pre) vs. 11.4% (post), p = 0.87.
Conclusion: Rapid Gram-negative microarray testing on positive blood cultures combined with real-time antimicrobial stewardship intervention was associated with decreased time to antimicrobial switch, time to active therapy, and hospital length of stay.
S. Lam, None
S. Gordon, None
G. Procop, None
S. S. Richter, Nanosphere: Research funding , Research support
bioMerieux: Research funding , Research support
BD Diagnostics: Research funding , Research support
Roche: Research funding , Research support
Biofire: Research funding , Research support
E. Neuner, None